Information Technology (IT) systems evolve with time; they are often maintained and upgraded by a separate department (or departments) from those deciding future enterprise directions and usages of the underlying system. This presents a challenge for enterprises and organizations wanting to construct an accurate model of processes, such as enterprise processes, corresponding to the currently-running IT systems (when such models do not exist), or wanting to track the changes or check the compliance of the IT systems (when the underlying processes do exist).
There are currently some techniques that engage in data mining from a system log, however, it is believed that these techniques do not focus on correlating user activity (via network and system monitoring) to the corresponding process. They also do not focus on automatic identification of the role players who act on a task or a collection of tasks in the process.
It would thus be desirable to overcome the limitations in previous approaches.